Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/miomio.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urllib_request
   8 from ..utils import (
   9     xpath_text,
  10     int_or_none,
  11     ExtractorError,
  12 )
  13
  14
  15 class MioMioIE(InfoExtractor):
  16     IE_NAME = 'miomio.tv'
  17     _VALID_URL = r'https?://(?:www\.)?miomio\.tv/watch/cc(?P<id>[0-9]+)'
  18     _TESTS = [{
  19         # "type=video" in flashvars
  20         'url': 'http://www.miomio.tv/watch/cc88912/',
  21         'md5': '317a5f7f6b544ce8419b784ca8edae65',
  22         'info_dict': {
  23             'id': '88912',
  24             'ext': 'flv',
  25             'title': '【SKY】字幕 铠武昭和VS平成 假面骑士大战FEAT战队 魔星字幕组 字幕',
  26             'duration': 5923,
  27         },
  28     }, {
  29         'url': 'http://www.miomio.tv/watch/cc184024/',
  30         'info_dict': {
  31             'id': '43729',
  32             'title': '《动漫同人插画绘制》',
  33         },
  34         'playlist_mincount': 86,
  35         'skip': 'This video takes time too long for retrieving the URL',
  36     }, {
  37         'url': 'http://www.miomio.tv/watch/cc173113/',
  38         'info_dict': {
  39             'id': '173113',
  40             'title': 'The New Macbook 2015 上手试玩与简评'
  41         },
  42         'playlist_mincount': 2,
  43     }]
  44
  45     def _real_extract(self, url):
  46         video_id = self._match_id(url)
  47         webpage = self._download_webpage(url, video_id)
  48
  49         title = self._html_search_meta(
  50             'description', webpage, 'title', fatal=True)
  51
  52         mioplayer_path = self._search_regex(
  53             r'src="(/mioplayer/[^"]+)"', webpage, 'ref_path')
  54
  55         http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path}
  56
  57         xml_config = self._search_regex(
  58             r'flashvars="type=(?:sina|video)&amp;(.+?)&amp;',
  59             webpage, 'xml config')
  60
  61         # skipping the following page causes lags and eventually connection drop-outs
  62         self._request_webpage(
  63             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/xml.php?id=%s&r=%s' % (id, random.randint(100, 999)),
  64             video_id)
  65
  66         vid_config_request = compat_urllib_request.Request(
  67             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/sina.php?{0}'.format(xml_config),
  68             headers=http_headers)
  69
  70         # the following xml contains the actual configuration information on the video file(s)
  71         vid_config = self._download_xml(vid_config_request, video_id)
  72
  73         if not int_or_none(xpath_text(vid_config, 'timelength')):
  74             raise ExtractorError('Unable to load videos!', expected=True)
  75
  76         entries = []
  77         for f in vid_config.findall('./durl'):
  78             segment_url = xpath_text(f, 'url', 'video url')
  79             if not segment_url:
  80                 continue
  81             order = xpath_text(f, 'order', 'order')
  82             segment_id = video_id
  83             segment_title = title
  84             if order:
  85                 segment_id += '-%s' % order
  86                 segment_title += ' part %s' % order
  87             entries.append({
  88                 'id': segment_id,
  89                 'url': segment_url,
  90                 'title': segment_title,
  91                 'duration': int_or_none(xpath_text(f, 'length', 'duration'), 1000),
  92                 'http_headers': http_headers,
  93             })
  94
  95         if len(entries) == 1:
  96             segment = entries[0]
  97             segment['id'] = video_id
  98             segment['title'] = title
  99             return segment
 100
 101         return {
 102             '_type': 'multi_video',
 103             'id': video_id,
 104             'entries': entries,
 105             'title': title,
 106             'http_headers': http_headers,
 107         }